1
Librarians and Classification: Defining a Middle Space of Social Responsibility
By Glenda Claborne
Between the individual and society
Can the social responsibility of librarians be located in some middle space between the individual and society? The difficulty of linking the individual and society has been a long-standing concern in sociology and this difficulty has again come to my attention in my recent readings on systems of classification. In particular is the book Sorting Things Out: Classification and Its Consequences by Geoffrey Bowker and Susan Starr (1999). The central argument of the book is that our systems of classification and standards create a social and moral order that have profound effects on the lives of individuals, work practices, and large-scale infrastructures. The authors point out that most of the workings of these classifications and standards are embedded in a dense network of cognitive, social, political, and technological structures. But they remain invisible to us and the authors implore us to take up, as our moral and ethical responsibility, the task of uncovering and mapping this territory. Reviewers of the book lauded Bowker’s and Starr’s efforts to make us recognize this responsibility but have criticized the book for failing to connect their study to the large body of existing cognitive research in classification (Brooks, 2000) and to the role of individuals in creating and using classifications (Beghtol, 2001).
That there seems to be a gulf difficult to span between the individual and society, at least in our theoretical analyses, can be seen in Durkheim’s and Mauss’ classic book Primitive Classification and in Rodney Needham’s introduction to his translation (from the French) of this book. In Primitive Classification, as in his later works, Durkheim’s concerns about the nature and origins of moral and religious order center on society itself. Society itself, not any innate capacity in the human mind, Durkheim argues, is the model for systems of categories and classifications. But Needham points out that a lot of a priori cognitive capabilities have already been presumed by Durkheim’s and Mauss’ concessions of individuals’ abilities to distinguish left from right and past from present, to perceive resemblances, to separate the one from the many, and to group things. These abilities, Needham argues, already presuppose formidable concepts of time, space, quantity, and quality in the human mind.
We maybe justified in declaring these omissions in analysis as merely difficulties in methodologies as defined by particular disciplines. Surely, we must all recognize the complexity of classifications and any attempt to give a comprehensive account of them, whether coming from the individual or society as a starting point, would inevitably obscure other aspects of the problem. But the fact remains that we all do acts of classification in our everyday lives, that some classifications become systematized and embedded in the infrastructures of society, and that the challenge remains to understand more about them so we may be able to recognize the dangers as well as the advantages that they carry.
In this essay, I take up the challenge with a focus on the advantages. It is my perception that whenever classification is considered in the realm of the social, we tend to emphasize its dangers as manifested in discrimination, exclusion, stereotypes, bias, and prejudice more than its advantages in helping us to simplify our environment, reduce the load on our memory, store and retrieve information efficiently, and generate new knowledge. I would like to go over these advantages and perhaps help define a middle space between social structures and cognitive structures, a space where librarians and information professionals can define their social responsibility vis-à-vis classification systems more effectively.
The advantages of classification
1. Classifications simplify our environment. When we divide up the universe of knowledge into classes or facets, we are sorting information objects into groups and into a system of relationships. This system helps us reduce chaos and confusion among the things that surround us. This highlights all the more the importance of classification systems as a way of dealing with information overload in the Information Age.
2. Classifications reduce the load on our memory. Imagine having to remember the features and properties of entities and events as if they are purely distinct individuals unrelated to one another. We would have to deal with each and every instance as if we were encountering them for the first time. Fortunately, classifications provide a way for us to make explicit the natural groupings that we see in the world as well as those artificial groupings that we create to understand complexity better.
3. Classification helps us store and retrieve information. This one is obvious enough to librarians and information professionals. The categories that we create become storage units into which we assign information resources and we use these units to name or label these objects so that we have a way of knowing where to retrieve them for use later.
4. Classification helps us generate new knowledge. The storage and retrieval functions of classification tend to obscure the fact that our systems of classification become information in their own right. I would guess that many people have used library catalogs not only to find information but to use it also like a reference tool, which could show the relationships between topics, between authors, between works, etc. From these gleanings, we come to know what is there and what is not there and perhaps induce us to create what is not there yet.
So how do we leverage these advantages of classification while at the same time be vigilant of the fact that, once systematized, classification systems acquire an inertia that indeed, over time, become discriminatory, exclusionary, and biased and that would take much more than a Sanford Berman to break? I suggest that we take a look at some research already done in our own field in library and information science that use existing research in sociology, linguistics, anthropology, cognitive psychology, and computer science. A review of these might help us find a fruitful middle ground between individual and social factors in the creation and use of classification systems.
Folk classification and basic level terms
In her article pointing out the human, database, and domain factors that can be utilized in the design of information systems, Marcia Bates (1997) pointed to research in linguistics and anthropology which show that across many cultures and languages, there is a common generic level consisting of a range of terms that correspond to groupings or discontinuities that are easily recognizable in nature. She notes the observation that the Linnaean system of classification in biology resembles this folk classification in its hierarchical structure and further notes that she wouldn’t be surprised if the Dewey Decimal system also exhibits this correspondence with folk classification. The important point that Bates was trying to point out here is that if we observe the pattern that most people from most cultures use a limited range of terms at a generic level in their information-seeking behaviors, then we have here a rich empirical pattern on which to model our information systems.
Bates also points to research in psychology where it was found that people employ a set of “basic terms” which also reflects groupings based on natural discontinuities. These basic terms are found to be neither the broadest nor the narrowest but at some level correspond to the generic level in folk classification. Again, if we observe the same pattern of language use among our library users, how will that affect our designs of catalogs, indexes, and controlled vocabularies?
Grounded theory and faceted classifications
Changes in the infrastructure of information retrieval and use and the nature of research has motivated Susan Leigh Starr (1998) to take Glaser’s and Strauss’ grounded theory as used in sociology and anthropology and compares and contrasts this theory with Ranganathan’s faceted classification. Starr sees a rich possibility of cross-fertilization between the concerns of each theory - the problems of classificationists in how to combine different vernacular and representational schemes or formal and informal schemes of representation and the concerns of grounded theorists about the quality of data analysis and theory development within the fluid, in-flux nature of the networked environment of the Information Age. These theories struggled with the core problem of how to represent both the local, specific, or empirical as well as the general and abstract in an integrated system of representation. Ranganathan’s faceted classification was in part a reaction to the limitations of then existing classification schemes to accommodate diverse, disparate knowledge. Similarly, Glaser and Strauss grounded theory sought to reform the strictures of quantitative methods of research in the social sciences.
Starr points out the rich structural properties of classification systems, particularly faceted classification, which can be used for data analysis and theoretical development as sought by grounded theorists. She mentions Kwasnik’s work in using Ranganathan’s faceted classification scheme to assess the construction of other classification schemes such as the periodic table of elements, the classification of psychiatric diseases in the DSM-III, and the classification of software re-use. How much do we know of the structural properties of the classification schemes that we use in library and information science and how may we leverage these to also help us in developing stronger theoretical foundations for our practices?
Constructing classificatory spaces
We find classification schemes as boundary systems designed to make explicit the discontinuities that we observe in the objects we are trying to organize but as boundary systems, we are also aware of their power to shut out minority voices. Hope Olson (1998) is realistic about the nature of classification schemes as social constructs that reflect the biases of their creators as they are products of their own times and cultures. However, Olson sought to find a way to open up boundaries in DDC, both as a critique and a tool for change, by borrowing theoretical frameworks from other disciplines in order to construct one that will adequately be used to analyze the spatial structure of classification schemes. Olson discusses other mapping tools such as Zipf’s, Lotka’s, and Bradford, which show a core concentration of mainstream concepts, surrounded by concentric circles of marginalized concepts. But Olson notes that these tools are inadequate in that they merely describe present arrangements and do not illuminate the historical discourses that have shaped the distributions. Furthermore, she notes that these tools do not offer suggestions on how to change the status quo. She turns to spatial metaphors as powerful tools for mapping marginalized knowledge domains, using terminology in A Woman’s Thesaurus as an example, to the mainstream domains represented in the DDC.
The process of mapping using spatial metaphors is not a process of shoehorning new spaces into old, rigid ones but of constructing “paradoxical spaces,” a term Olson borrowed from feminist geographer Gillian Rose. Olson gives Rose’s definition of paradoxical space as “simply a practice that allows existence of a limit simultaneously or alternately. It is both inside and outside, center and margins.” A paradoxical space “puts a different spin on existing concepts that come to co-exist with concepts from the margins.” In what other ways can we visualize classification schemes and find in its structures not just boundaries that shut out minority from mainstream concepts but openings for combination and fruitful co-existence?
Conclusion
I have set out in this essay with a question that implies a dichotomy between the individual and society but with a hope of finding a middle space in which we can exercise our social responsibility as librarians in view of the power of classifications to unite, divide, and marginalize universes of knowledge and the populations that use them. But a review of some research on the issues posed by classification shows that my question is simplistic. The issues go beyond the relationship between society and the individual, between social structures and cognitive structures. The network of variables that define classifications is as complex as the universes of knowledge that they are trying to organize.
The research that I have looked at show us ways in which we can leverage the advantages of classifications especially if we consider that classifications are perhaps one of the greatest achievements of the human mind as well as societies. There are ways to overcome the limits, the inadequacies, and the boundaries of our classification schemes. These ways towards change require us indeed to be more aware of studies done in other disciplines and that have potential to illuminate our own problems in LIS, in both theory and practice. It is also refreshing to see that the structures of classification schemes in library and information science have rich potential as analytical tools for other disciplines.
One of the concerns that have led me to pose the question at the beginning of this essay was where to locate our passions for social responsibility as librarians. I think I have feared that a middlespace would only mean debilitating neutrality. What I found instead is a challenging space in which my capacities as a future librarian and information professional can be grounded, expanded, and applied to its fullest. My social responsibility as a librarian is to be well-informed of what’s going on in other domains of knowledge so that I have a strong knowledge of what needs to be changed and how. On this foundation, I believe that I can be more effective in my service to individuals and societies.